Investigating backtranslation for the improvement of English-Irish machine translation
In this paper, we discuss the difficulties of building reliable machine translation systems for the English-Irish (EN-GA) language pair. In the context of limited datasets, we report on assessing the use of backtranslation as a method for creating artificial EN-GA data to increase training data for...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
The Irish Association for Applied Linguistics
2019-11-01
|
| Series: | Teanga: The Journal of the Irish Association for Applied Linguistics |
| Subjects: | |
| Online Access: | https://journal.iraal.ie/index.php/teanga/article/view/88 |
| _version_ | 1818261949917429760 |
|---|---|
| author | Meghan Dowling Teresa Lynn Andy Way |
| author_facet | Meghan Dowling Teresa Lynn Andy Way |
| author_sort | Meghan Dowling |
| collection | DOAJ |
| description | In this paper, we discuss the difficulties of building reliable machine translation systems for the English-Irish (EN-GA) language pair. In the context of limited datasets, we report on assessing the use of backtranslation as a method for creating artificial EN-GA data to increase training data for use state-of-the-art data-driven translation systems. We compare our results to earlier work on EN-GA machine translation by Dowling et al (2016, 2017, 2018) showing that while our own systems do not compare in quality with respect to traditionally reported BLEU metrics, we provide a linguistic analysis to suggest that future work with domain specific data may prove more successful. |
| first_indexed | 2024-12-12T18:55:22Z |
| format | Article |
| id | doaj.art-38299098bf4c45b88d8a8f498e67713e |
| institution | Directory Open Access Journal |
| issn | 0332-205X 2565-6325 |
| language | English |
| last_indexed | 2024-12-12T18:55:22Z |
| publishDate | 2019-11-01 |
| publisher | The Irish Association for Applied Linguistics |
| record_format | Article |
| series | Teanga: The Journal of the Irish Association for Applied Linguistics |
| spelling | doaj.art-38299098bf4c45b88d8a8f498e67713e2022-12-22T00:15:15ZengThe Irish Association for Applied LinguisticsTeanga: The Journal of the Irish Association for Applied Linguistics0332-205X2565-63252019-11-012610.35903/teanga.v26i0.88Investigating backtranslation for the improvement of English-Irish machine translationMeghan Dowling0Teresa Lynn1Andy Way2ADAPT Centre, Dublin City UniversityADAPT Centre, Dublin City UniversityADAPT Centre, Dublin City UniversityIn this paper, we discuss the difficulties of building reliable machine translation systems for the English-Irish (EN-GA) language pair. In the context of limited datasets, we report on assessing the use of backtranslation as a method for creating artificial EN-GA data to increase training data for use state-of-the-art data-driven translation systems. We compare our results to earlier work on EN-GA machine translation by Dowling et al (2016, 2017, 2018) showing that while our own systems do not compare in quality with respect to traditionally reported BLEU metrics, we provide a linguistic analysis to suggest that future work with domain specific data may prove more successful.https://journal.iraal.ie/index.php/teanga/article/view/88Machine translationBacktranslationIrish languageAutomatic Evaluation Metrics |
| spellingShingle | Meghan Dowling Teresa Lynn Andy Way Investigating backtranslation for the improvement of English-Irish machine translation Teanga: The Journal of the Irish Association for Applied Linguistics Machine translation Backtranslation Irish language Automatic Evaluation Metrics |
| title | Investigating backtranslation for the improvement of English-Irish machine translation |
| title_full | Investigating backtranslation for the improvement of English-Irish machine translation |
| title_fullStr | Investigating backtranslation for the improvement of English-Irish machine translation |
| title_full_unstemmed | Investigating backtranslation for the improvement of English-Irish machine translation |
| title_short | Investigating backtranslation for the improvement of English-Irish machine translation |
| title_sort | investigating backtranslation for the improvement of english irish machine translation |
| topic | Machine translation Backtranslation Irish language Automatic Evaluation Metrics |
| url | https://journal.iraal.ie/index.php/teanga/article/view/88 |
| work_keys_str_mv | AT meghandowling investigatingbacktranslationfortheimprovementofenglishirishmachinetranslation AT teresalynn investigatingbacktranslationfortheimprovementofenglishirishmachinetranslation AT andyway investigatingbacktranslationfortheimprovementofenglishirishmachinetranslation |